Transit reliability is more than a number. It’s whether your bus shows up when the app says it will, whether your train skips your stop without warning, and whether you can reliably make work, school, or daycare on time. This Transit On-Time Report explains how agencies measure punctuality, why cancellations and “ghost buses” happen, what fixes actually improve reliability, and how riders can reduce day-to-day stress while agencies upgrade the system.
What “on-time” really means
Most agencies define on-time as a vehicle arriving no more than a certain number of minutes early or late.
- Schedule adherence (typical for commuter rail and timed bus routes)
- On time = 0 minutes early to 5 minutes late (varies by agency)
- Early departures are penalized because they strand riders
- Headway adherence (typical for high-frequency bus/tram/metro)
- Goal is even spacing (e.g., a bus every 6–10 minutes)
- A vehicle can be “on time” even without a fixed timetable if gaps and bunching are minimized
Key takeaway: Headway-based lines should be measured by spacing, not timetables. Riders experience gaps and bunching as unreliability—even if the average is “on time.”
The reliability metrics that matter
- On-time percentage: Share of trips meeting schedule or headway targets
- Excess wait time (EWT): How much longer riders wait compared to the planned frequency
- Missed trips/cancellations: Scheduled trips that never run (crew shortage, equipment failure)
- Short turns/skips: Trips that end early or skip stops to “get back on time”
- Real-time accuracy: How often the countdown matches actual arrivals (ghost buses/trains erode trust)
- Crowding: Pass-ups and standee load—reliability includes the ability to board
Why buses and trains run late—or not at all
Common causes by mode
- Bus
- Traffic congestion, double parking, and blocked stops
- Long dwell times from slow boarding and fare payment
- Construction detours without schedule adjustments
- Irregular dispatching and holding practices
- Rail (metro/commuter)
- Signal and switch failures; power issues
- Door holding and extended dwell at crowded stations
- Track maintenance and speed restrictions
- Crew and equipment shortages
Back of house matters
- Fleet condition and spare ratio: Fewer spare vehicles = more cancellations when a few break
- Crew availability: Tight staffing leads to cut trips, especially early morning and late night
- Schedule design: Not enough recovery time at terminals guarantees late departures later in the day
- Real-time systems: Inaccurate AVL/vehicle tracking creates ghost vehicles and bad countdowns
Cancellations and “ghost buses”: what’s going on?
- Cancellations: A scheduled trip doesn’t leave the depot. Causes include crew absence, mechanical failure, or planned reductions without timely public notice.
- Ghost buses/trains: Real-time apps show a vehicle approaching, then it disappears or never arrives. Causes include incorrect GPS, last-minute short turns, or data not reflecting operational changes.
Rider impact: Ghosting is worse than a posted delay because it undermines trust and forces “just-in-case” buffer time.
The fixes that actually improve reliability
Bus priorities and faster boarding
- Dedicated bus lanes and queue jumps: Keep buses moving around bottlenecks
- Transit Signal Priority (TSP): Extends green or shortens red for buses approaching intersections
- All-door boarding and off-board fare payment: Cuts dwell time dramatically on busy corridors
- Stop balancing: Removing closely spaced, low-ridership stops speeds trips and stabilizes headways
- Platform upgrades: Level boarding and near-level curbs reduce boarding time for everyone
Smarter scheduling and operations
- Headway management: Use real-time holding and dynamic dispatch to even out spacing
- Terminal recovery time: Build enough layover to absorb variability without cascading delays
- Crew and fleet resilience: Increase spare ratio and cross-qualification to avoid cancellations
- Manage demand: Adjust frequency to match peaks; add short-turn trips where crowding is chronic
- Accurate detour planning: Publish adjusted schedules and stop locations for construction/closures
Rail reliability
- State of good repair: Fund signals, power, and track to reduce service outages
- Platform staff at hotspots: Speed boarding, reduce door holds, and assist accessibility
- Turnback efficiency: Use center tracks and pocket tracks to recover from disruptions without systemwide delays
- Clear incident comms: Fast, honest updates and alternatives limit knock-on delay
Transparency and data riders can use
- Open data (GTFS + GTFS-RT): Agencies should publish both schedules and real-time feeds so apps can reflect true conditions
- Accuracy audits: Public dashboards should report real-time accuracy, not just scheduled on-time
- Standard reporting:
- On-time by route and time of day
- Cancellations/missed trips
- Excess Wait Time for frequent routes
- Crowding and pass-up rates
- Connection protection: Show when buses/trains will be held for late connections and where
How to read your agency’s on-time dashboard
Look beyond the headline percentage.
- Frequent routes: Check EWT and headway irregularity (gaps over 2x planned headway)
- Peak vs. off-peak: Some lines do well mid-day but collapse in the peak (or vice versa)
- Directional differences: Inbound mornings and outbound evenings may have distinct issues
- Top five pinch points: Intersections, terminals, or track segments with repeat delays—see if fixes are planned
Rider playbook: reduce day-to-day uncertainty
Trip planning
- Use two sources: Check your agency’s app plus a third-party app for redundancy
- Bookmark service alerts: Construction, detours, and special events can change stops/times
- Identify frequency zones: On frequent routes, plan for headway-based waits rather than fixed times
En route
- Favor stops with multiple routes: Flexibility increases your chances of a quick ride
- Mind the first and last mile: A short walk to a faster corridor often beats waiting for a slow local
- Connection buffer: Build a 5–10 minute buffer for critical transfers or use guaranteed transfer points when offered
Accessibility and safety
- Level boarding stops save time and reduce falls; report broken ramps or obstructed stops
- Night travel: Choose well-lit stops; use “get off at next stop” features where available
What riders can ask agencies to prioritize
- Bus lanes and TSP on the worst 10% of corridors for delay
- All-door boarding and off-board payment on high-ridership lines
- Reliable headway management with visible standards (e.g., ≤2 minutes variance on 6–10 minute routes)
- Real-time accuracy targets (e.g., >90% arrivals within 1 minute of countdown) and ghost-bus audits
- Clear cancellation reporting: Why it happened and what’s changing
- Connection protection at major hubs during off-peak periods
- Equity reporting: Reliability by neighborhood and time of day, not just systemwide averages
- Paratransit on-time and trip denials published alongside fixed-route metrics
Case study patterns: what works in practice
- Bus lanes + TSP: 10–25% travel time improvements on busy corridors, fewer bunching episodes
- All-door boarding: 25–50% dwell time reductions at major stops, higher schedule adherence
- Stop balancing: 5–15% running time savings with minimal ridership impact when community engagement is strong
- Frequent service with headway control: Lower excess wait time and fewer missed connections than timetable-driven operations
Cancellations, refunds, and rider rights
- Missed trip policies: Some agencies offer fare credits for significant delays or cancellations—know the thresholds
- Fare capping: Protects riders who are forced to re-tap due to ghost vehicles or skipped stops
- Service guarantees: Commuter rail often offers partial refunds for monthly pass holders when on-time falls below target
Frequently asked questions
Why is my bus always bunched?
Without headway management, minor delays compound: one bus gets crowded (longer dwell), the next catches up (emptier), and spacing collapses. Real-time holding and bus priority break the cycle.
Why do apps show a bus that never comes?
Ghost buses occur when vehicle data is wrong, a trip short-turns, or a cancellation isn’t pushed to the feed. Agencies should audit GTFS-RT accuracy and suppress bad predictions quickly.
Do bus-only lanes really help?
Yes. They reduce random delay from traffic, making schedules predictable. Combined with TSP and faster boarding, they deliver the biggest reliability gains per dollar.
Why not just add more recovery time?
Padding helps, but too much slows everyone and can hide deeper issues. Balance recovery with priority treatments and headway control for lasting improvement.
What’s the difference between “on time” and “reliable”?
On time is a threshold; reliability is how consistently service meets rider expectations—spacing, crowding, and real-time accuracy included.
A rider-focused reliability checklist
- On frequent lines, target headway variance, not just schedule adherence
- Publish and fix ghost-bus rates; set real-time accuracy targets
- Build bus lanes/TSP where delay is worst; add all-door boarding on top routes
- Balance stops and add level boarding to cut dwell and improve accessibility
- Provide connection protection and clear incident comms
- Report cancellations transparently with corrective actions
- Measure fairness: reliability by neighborhood/time, not just overall averages
The bottom line
Reliable transit is a design choice. Agencies that measure what riders feel—gaps, bunching, cancellations, and real-time accuracy—can target fixes that work: bus lanes, signal priority, faster boarding, right-sized schedules, and active headway management. Riders can cut uncertainty with smarter planning and by pushing for the proven playbook. When the countdown clock matches reality and vehicles arrive evenly and consistently, “on time” becomes more than a metric—it becomes your daily experience.
